On Tue, Jan 03, 2006 at 01:30:47PM -0500, bfields wrote: > As implementers of the Linux NFSv4 server support, we've found there's > interest in good support for NFS exports of cluster filesystems (via > NFSv4 and earlier versions). > > There are a number of obstacles to this, and we're interested in finding > solutions that are acceptable to OCFS2. (If I've directed this to the > wrong email list, please let me know!) > > To give an example--there are a couple of problems with the current VFS > support for posix byte-range locks:
I noticed just the other day that OCFS2 had popped up in linux/fs/ the other day. Yay! On a quick skim it doesn't appear to define a ->lock() method anywhere. Which I assume means that it doesn't attempt to provide cluster coherent posix byte-range (fcntl/lockf) locks. Are there plans to change this? --b. _______________________________________________ Ocfs2-devel mailing list [email protected] http://oss.oracle.com/mailman/listinfo/ocfs2-devel
